Abstract
We propose and evaluate a systematic approach to detect and classify Patient/Problem, Intervention, Comparison and Outcome (PICO) from the medical literature. The training and test corpora were generated systematically and automatically from structured PubMed abstracts. 23,472 sentences by exact pattern match of head words of P-I-O categories. Afterward, the terms with top frequencies were used as the features of Naïve Bayesian classifier. This approach achieves F-measure values of 0.91 for Patient/Problem, 0.75 for Intervention and 0.88 for Outcome, comparable to previous studied based on mixed textural, paragraphical, and semantic features. In conclusion, we show that by stricter pattern matching criteria of training set, detection and classification of PICO elements can be reproducible with minimal expert intervention. The results of this work are higher than previous studies.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 2011 IEEE International Conference on Granular Computing, GrC 2011 |
| Pages | 279-283 |
| Number of pages | 5 |
| DOIs | |
| Publication status | Published - 2011 |
| Event | 2011 IEEE International Conference on Granular Computing, GrC 2011 - Kaohsiung, Taiwan Duration: Nov 8 2011 → Nov 10 2011 |
Other
| Other | 2011 IEEE International Conference on Granular Computing, GrC 2011 |
|---|---|
| Country | Taiwan |
| City | Kaohsiung |
| Period | 11/8/11 → 11/10/11 |
Fingerprint
Keywords
- information extraction
- natural language processing
- question answering
- text mining
ASJC Scopus subject areas
- Software
Cite this
Classification of PICO elements by text features systematically extracted from PubMed abstracts. / Huang, Ke Chun; Liu, Charles Chih Ho; Yang, Shung Shiang; Xiao, Furen; Wong, Jau Min; Liao, Chun Chih; Chiang, I. Jen.
Proceedings - 2011 IEEE International Conference on Granular Computing, GrC 2011. 2011. p. 279-283 6122608.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Classification of PICO elements by text features systematically extracted from PubMed abstracts
AU - Huang, Ke Chun
AU - Liu, Charles Chih Ho
AU - Yang, Shung Shiang
AU - Xiao, Furen
AU - Wong, Jau Min
AU - Liao, Chun Chih
AU - Chiang, I. Jen
PY - 2011
Y1 - 2011
N2 - We propose and evaluate a systematic approach to detect and classify Patient/Problem, Intervention, Comparison and Outcome (PICO) from the medical literature. The training and test corpora were generated systematically and automatically from structured PubMed abstracts. 23,472 sentences by exact pattern match of head words of P-I-O categories. Afterward, the terms with top frequencies were used as the features of Naïve Bayesian classifier. This approach achieves F-measure values of 0.91 for Patient/Problem, 0.75 for Intervention and 0.88 for Outcome, comparable to previous studied based on mixed textural, paragraphical, and semantic features. In conclusion, we show that by stricter pattern matching criteria of training set, detection and classification of PICO elements can be reproducible with minimal expert intervention. The results of this work are higher than previous studies.
AB - We propose and evaluate a systematic approach to detect and classify Patient/Problem, Intervention, Comparison and Outcome (PICO) from the medical literature. The training and test corpora were generated systematically and automatically from structured PubMed abstracts. 23,472 sentences by exact pattern match of head words of P-I-O categories. Afterward, the terms with top frequencies were used as the features of Naïve Bayesian classifier. This approach achieves F-measure values of 0.91 for Patient/Problem, 0.75 for Intervention and 0.88 for Outcome, comparable to previous studied based on mixed textural, paragraphical, and semantic features. In conclusion, we show that by stricter pattern matching criteria of training set, detection and classification of PICO elements can be reproducible with minimal expert intervention. The results of this work are higher than previous studies.
KW - information extraction
KW - natural language processing
KW - question answering
KW - text mining
UR - http://www.scopus.com/inward/record.url?scp=84863028228&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84863028228&partnerID=8YFLogxK
U2 - 10.1109/GRC.2011.6122608
DO - 10.1109/GRC.2011.6122608
M3 - Conference contribution
AN - SCOPUS:84863028228
SN - 9781457703713
SP - 279
EP - 283
BT - Proceedings - 2011 IEEE International Conference on Granular Computing, GrC 2011
ER -